Home | Giacomo Camposampiero Personal Website

Hi, I'm Giacomo 👋🏻

I am a predoctoral researcher at IBM Research and a doctorate student in the DISCO group at ETH Zürich. I work on compositionality and systematic generalization, with the goal of improving the reasoning capabilities, sample efficiency, and robustness of neural models. Previously, I obtained my MSc in Computer Science from ETH Zürich and my BSc in Computer Engineering from the University of Padua. If you want to get in touch with me, reach out on social media or send me an email.

/ name [dot] surname [at] hotmail.com /
github /
linkedin /
scholar /

Selected Publications 📚

See all publications

Scalable Evaluation and Neural Models for Compositional Generalization

G. Camposampiero, P. Barbiero, M. Hersche, R. Wattenhofer and A. Rahimi

NeurIPS, 2025
I-RAVEN-X: Benchmarking Generalization and Robustness of Analogical and Mathematical Reasoning in Large Language and Reasoning Models

G. Camposampiero, M. Hersche, R. Wattenhofer, A. Sebastian and A. Rahimi

MATH-AI @ NeurIPS, 2025
Kernel approximation using analogue in-memory computing

J. Büchel, G. Camposampiero, A. Vasilopoulos, C. Lammie, M. Le Gallo, A. Rahimi and A. Sebastian

Nature Machine Intelligence, 2024
Towards Learning Abductive Reasoning using VSA Distributed Representations

G. Camposampiero, M. Hersche, A. Terzic, R. Wattenhofer, A. Sebastian and A. Rahimi

NeSy, 2024 [Spotlight]
Limits of Transformer Language Models on Learning to Compose Algorithms

J. Thomm, G. Camposampiero, A. Terzic, M. Hersche, B. Schölkopf and A. Rahimi

NeurIPS, 2024

Latest posts ✏️

See all posts

Contrastive learning, in short

A gentle introduction to the magical world of contrastive learning.